52 JOURNAL OF CONCHOLOGY, VOL. I3, NO. 2, APRIL, I9IO. 



Assiduous working of the locality will reveal a number of immature 

 shells, still very small, and perhaps isolated examples of the Alter- 

 nate type, which latter we may presuppose to be tiie direct descend- 

 ants of the former generation of Alternates. (This period in the 

 phase is indicated in the example given by an arrow, thus ■^). A 

 year later the big form again makes its appearance, reproduction takes 

 place, and so the history repeats itself. 



Thus we have in these cases of Revertive Alternation a period of 

 three years elapsing between each appearance of the major form. 



C. — Arrested gradation. 



It is not possible at the present to do more than suggest the nature 

 of this phase. There are localities at which a large ovate form of 

 pereger makes a regular appearance each year. Once every five or 

 six years a maximum size is seen to be attained, to be frequently 

 followed by a considerable falling off the following year. It is prob- 

 able that at these localities the circumstances tavourable to extreme 

 variation have reached the optimum. It seems fairly evident that 

 both Alternation and Reversion are present during the same period, 

 and in this connection it should be noted that there is ample proof 

 that cross-breeding between the two generations takes place. With 

 these facts alone, it will be seen that it becomes extremely difficult 

 to get conclusive evidence, and we can only take the facts obtained 

 and consider them in the light of the preceding phases. 



A final word as to the support given to these facts by the observa- 

 tions of other workers. 



In the Proceedings of the Woolhope Naturalists' Field Club, 

 Messrs. Boycott and Bowell instance the following cases of periodic 

 variation under Liinncea Iwiosa L. : — 

 "Owing to various causes we have not been able to give this 

 periodic variation sufficient study, but the general result of our 

 researches is to show that in a series of favourable seasons the 

 size of average specimens gradually increases till a maximum is 

 attained ; and then, as a rule, these large-sized specimens pro- 

 duce a very numerous progeny of small specimens, which in 

 their turn present us with offspring larger than themselves, and 

 so on. In a pond under observation, at Sissinghurst in Kent, 

 a biennial change has occurred with some regularity during the 

 last few years, a labiosa form producing large iiniosa, and these 

 in their turn reverting to labiosa in successive seasons." 



